An Analysis of Ambulance Response Times Across the State of Alabama

1

An Analysis of Ambulance Response Times Across the State of Alabama

Daniel M. KleinFaculty Sponsor: Dr. Allen Parrish

Center for Advanced Public Safety December 7, 2012

2 12/7/12 -- EMS Analysis

106,000

6,000

2,000 per year

• Alabama injury crashes, Jan. 2009 – Aug. 2012

• Injury crashes with ambulance delay over 30 minutes

3

An Analysis of Ambulance Response Times Across the State of Alabama

Daniel M. KleinFaculty Sponsor: Dr. Allen Parrish

Center for Advanced Public Safety December 7, 2012

4 12/7/12 -- EMS Analysis

Background• Concept proposed by Dr. David Brown of Center for

Advanced Public Safety (CAPS)

• Shocking frequency of extended delays

• Potential to save lives• Dataset: Alabama crashes from 2009 to August

2012• Best reflects modern reality

5 12/7/12 -- EMS Analysis

Research Goals

city-data.com

• Facilitate decision-making process• Extract actionable

information from data

• Illustrate with heat maps

6 12/7/12 -- EMS Analysis

Critical Analysis Reporting Environment

caps.ua.edu

• Data analysis software package

• Developed for traffic safety applications• Information mining capability – IMPACT

• Extract information from data

• Present results graphically and tabularly

7 12/7/12 -- EMS Analysis

Adjusted Emergency Medical Services Arrival Delay (AEMSAD)

• EMS Delay variable not optimal

• AEMSAD variable created for CARE warehouse

• Better reflects reality

• Helps extract valid information from imperfect data

al.com

8 12/7/12 -- EMS Analysis

IMPACT Analysis

• Designed AEMSAD GT 30 and LT 30 filters

• AEMSAD greater or less than 30 minutes

• Injury crashes involving EMS transport only

• Executed IMPACT on all variables

• Revealed variables correlated to higher

AEMSAD

9

IMPACT Results – Time of Day

12/7/12 -- EMS Analysis • 9

10 12/7/12 -- EMS Analysis

Geographic Information Systems (GIS) Analysis

esri.com

• Locational data exported from CARE

• Coordinate points with related variables

• Data processed with ArcGIS

• Heat map generation

11 12/7/12 -- EMS Analysis

Heat Map Generation Process

• Data exported from CARE into shapefile

• Added to CARE base map of Alabama

• Further data manipulation performed in ArcGIS

• Clean data, Integrate and Collect Events, and Spatial Join

• Finally create hot spots and interpolate

12 12/7/12 -- EMS Analysis

Heat Map Result

• Compares average

AEMSAD values for given

locations

• Red indicates statistically

higher delay, blue lower

13 12/7/12 -- EMS Analysis

Next Steps

caps.ua.edu

• Performing more pinpointed analysis

• Specific locales and routes

• Empowering officials with information

• Tactical and strategic decisions

14

Questions?

12/7/12 -- EMS Analysis

The University of Alabama Center for Advanced Public

SafetyTuscaloosa, AL

(205) 348-6999 http://caps.ua.edu

Allen Parrish, Director

[email protected]

Daniel [email protected]

(901) 378-4678

Special thanks to project sponsor Dr. Allen Parrish,

Dr. David Brown, Dr. Randy Smith, and Jesse Norris of CAPS for their guidance on

this project.

15 12/7/12 -- EMS Analysis

Works Cited"Center for Advanced Public Safety." Center for Advanced Public

Safety. The University of Alabama, n.d. Web. Oct.-Nov. 2012.

<http://caps.ua.edu/>.

National Association of State EMS Officials. Proof of Concept for a

Nationwide Highway Mass Casualty Readiness Measurement

Project: Model Inventory of Emergency Care Elements “MIECE”. Mar.

2011.

Norris, Jesse, and David Brown. Analysis of the Most Critical Factors

in Teen (15-19) Driver Causes Vehicle Crashes. 17 Nov. 2011.

16 12/7/12 -- EMS Analysis

Learning CARE

• Filter creation

• Understanding logic

• Frequency analysis

• Locational analysis and hot spots caps.ua.edu

17 12/7/12 -- EMS Analysis

AEMSAD Definition• If Police Notification Delay < 10 minutes OR EMS Arrival

Time prior to Police Arrival Time:• AEMSAD = EMS Delay (standard)

• If Police Notification Delay > 10 minutes AND EMS Arrival Time after Police Arrival Time:• AEMSAD = EMS Arrival Time – Police Notification Time

18 12/7/12 -- EMS Analysis

About IMPACT

• Information mining by comparing two subsets

• Performed on all variables to extract information

• Graphical and tabular results

19 12/7/12 -- EMS Analysis

IMPACT Examplecaps.ua.edu

20 12/7/12 -- EMS Analysis

IMPACT Results – Police Arrival Delay

12/7/12 -- EMS Analysis • 20